Inhibition of E-selectin and VCAM-1, but not other adhesion molecules, attenuates vincristine-induced mechanical hypersensitivity and F4/80⁺ immune cell accumulation in the dorsal root ganglia (DRG) and sciatic nerves (SN).

The effect of blocking antibodies against VCAM-1 (VCAM1ab; A, B), ICAM-1 (ICAM1ab;C,D); PECAM-1 (E,F), P-selectin (P-sel ab; G,H) and E-selectin (E-sel ab; J,K) on vincristine-induced mechanical hypersensitivity (A,C,E,G,J; paw withdrawal threshold (g)) and the number of F4/80⁺ cell in dorsal root ganglia (DRG) and sciatic nerve (SN) (B,D,F,H,K; %DAB). Black arrows indicate injection (i.p.) of antibodies (10 mg/kg), isotype control antibodies (IgG2B, IgG1; (10 mg/kg)) and vincristine (0.5 mg/kg). Both VCAM-1 and E-selectin blocking antibodies alleviated vincristine-induced mechanical hypersensitivity (A,J) as well as the increase in F4/80⁺ cells in dorsal root ganglia (DRG) and sciatic nerve (SN) (B,K), while PECAM-1 and P-selectin blocking antibodies were partially effective (E-H) and ICAM-1 blocking antibody had not effect (C,D). Data are presented as mean ± SEM (n = 5–7 per group). Statistical significance (# p < 0.05) was determined using repeated-measures two-way ANOVA or one-way ANOVA with Šídák’s multiple comparisons test.

Vincristine induces mechanical hypersensitivity and F4/80⁺ macrophage accumulation, but not peripheral nerve damage, that is attenuated in Sele⁻/⁻ mice.

(A) Sele⁻/⁻ mice were protected from vincristine-induced mechanical hypersensitivity compared to wt C57BL6/J controls. Mice received vincristine (VCR) for five consecutive days (black arrows), and mechanical hypersensitivity was assessed at baseline (0 h) and on days 1, 2, 4 and 7 post-treatment. (B,C) In contrast to wt C57BL6/J mice, Sele⁻/⁻ mice treated with vincristine were protected from F4/80⁺ cell accumulation in the dorsal root ganglia (DRG) and sciatic nerve (SN) 25 h after the first vincristine administration. (D) Representative images of F4/80⁺ immunoreactivity in DRG and SN from wt C57BL6/J and Sele⁻/⁻ mice. Arrowheads indicate F4/80⁺ areas. Scale bars, 20 µm (DRG) and 10 µm (SN). (E-H) Vincristine does not cause structural damage to peripheral nerve fibres. No significant differences in intraepidermal nerve fibre (IENF) density (E), myelinated fibre density (F,G) or relative myelinated fibre size distribution in caudal nerves (H) were observed between saline and vincristine treatments in both wt C57BL6/J and Sele⁻/⁻ mice. (F) Representative images of myelinated fibres in caudal nerves of wt C57BL/6J and Sele⁻/⁻ mice treated with vincristine or saline. Scale Bar 10 µm. Statistical analyses were performed using repeated-measures two-way ANOVA with Sidak’s multiple-comparison test (A,H) or two-tailed t-tests (B–G). #: P < 0.05; n = 3–8 per group.

Vincristine reshapes cell-cell communication in dorsal root ganglia through injury-associated signalling pathways.

CellChat analysis was used to infer ligand–receptor–mediated communication networks in dorsal root ganglia (DRG) following vincristine treatment, revealing pathway-specific sender, receiver, mediator and influencer roles across neuronal and non-neuronal populations. Heatmaps (A,C,E,G,I,K,M,O) show the relative importance of each cell type as senders, receivers, mediators or influencers, with darker green indicating greater contribution. Network schematics (B,D,F,H,J,L,N,P) depict dominant ligand–receptor interactions (thickness of the connector line depicts the relative contribution of the L-R pair). (A-H) Analysis of Spp1 (A,B), galectin (C,D), Ptn (E,F) and Sema3 (G,H) signalling in vincristine-treated C57BL6/J mice. (I–P) Corresponding analyses in Sele⁻/⁻ mice treated with vincristine demonstrate preservation of the same signalling pathways but redistribution of communication roles, with changed immune-cell, stromal and glial populations participation as senders, mediators and influencers across Spp1, galectin, Ptn and Sema3 pathways. (Q) Dot plot showing average expression (colour scale) and proportion of expressing cells (dot size) for immune pathway-associated genes across all annotated cell types in Sele⁻/⁻ (VCR) mice. VCR: vincristine; Sele⁻/⁻: E-selectin–deficient mice; DRG: dorsal root ganglion; Schwann_M: myelinating Schwann cells; Calca⁺Bmpr1b and Calca⁺Smr2: peptidergic sensory neuron subtypes expressing Calca with Bmpr1b or Smr2, respectively; Mrgprd: non-peptidergic sensory neurons; Mrgpra3⁺Trpv1: nociceptive/pruriceptive neurons; Ntrk3^high/+Ntrk2 and Ntrk3^low/+Ntrk2: sensory neuron subsets defined by relative Ntrk3 and Ntrk2 expression; Pvalb: parvalbumin-expressing interneurons; Sst: somatostatin-expressing neurons; Th: tyrosine hydroxylase–expressing neurons; Immune: aggregated immune cell populations; Itgav/Itgb: integrin αV/β subunits; Nrp: neuropilin; Plxna: plexin A.

E-selectin-induced mechanical hypersensitivity is mediated by immune cells.

(A) Local intraplantar (i.pl.) injection of E-selectin – but not saline or IgG2b – into the hind paw of C57BL6/J mice induces significant mechanical hypersensitivity, (B) but not thermal hypersensitivity, and is (C) accompanied by an increase in F4/80⁺ cells in the injected footpads. (D) Representative images of F4/80⁺ cells in the E-selectin, IgG2b and saline-injected food pads. Scale bar = 10 µm. (E) Fut4/7⁻/⁻ mice are protected from E-selectin-induced mechanical hypersensitivity and (F) the E-selectin-induced increase in F4/80⁺ cell area. (G–H) Adhesion of dorsal root ganglion neurons (DRG) or bone marrow-derived macrophages (BMDMs) to E-selectin (E-sel), isotype control (IgG2b), or buffer. (G) E-selectin does not increase DRG neuron adhesion, but (H) significantly increases BMDM adhesion, which is significantly reduced by pre-treatment with an E-selectin–blocking antibody (E-sel Ab). (I) Representative immunohistochemical images of F4/80⁺ cells in the spleens of C57BL6/J mice pretreated with clodronate (CLD) or liposome control (LIP) and injected i.pl. with E-sel, demonstrating effective depletion of F4/80⁺ cells. Scale bar = 10 µm. (J) Pretreatment with liposomal clodronate (CLD) protects C57BL6/J mice from E-selectin-induced mechanical hypersensitivity. (K) CLD pretreatment also significantly reduces the increase in F4/80⁺ cell area in the footpad following i.pl. E-selectin injection. Statistical significance (#: P < 0.05) was determined using one-way ANOVA for C, F, G, H (n=3-9) or repeated measures two-way ANOVA for A, B, E, J,K (n = 6-9).

E-selectin adhesion specifically enhances interleukin-1β release in vincristine-treated BMDMs.

(A-G) Release of IL-1β (A), IL-18 (B), IL-10 (C), G-CSF (D), IL12p40 (E), IL-6 (F) and CXCL1 (G) is significantly higher in LPS-primed vincristine-treated (LPS+VCR) BMDMs compared to untreated (PBS), LPS-primed (LPS) or vincristine-treated (VCR) BMDMs. (H-N) Release of IL-18 (H), G-CSF (I), IL-12 (J), IL-6 (K) and CXCL1 (L) is not different between BMDMs adhered to IgG2b (control) or E-sel (E-selectin). Release of IL-10 is ∼2.5-fold higher (M) and release of IL-1β is ∼1.4-fold higher (N) in BMDMs adhered to E-selectin. Statistical significance was determined using one-way ANOVA with Dunnett’s multiple comparisons test, comparing LPS + VCR to LPS alone (A-G) or E-sel adhered to IgG2b adhered condition (n = 4 per group; #: P < 0.05).

E-selectin functions as a non-canonical signalling amplifier of NF-κB–dependent NLRP3 activation and IL-1β release, driving vincristine-induced mechanical hypersensitivity.

(A) Representative images of ASC specks (green; white arrows) in bone marrow-derived macrophages (BMDMs) treated with lipopolysaccharide (LPS, 10 ng/mL) and vincristine (VCR, 100 µM). Nuclei are stained with DAPI (blue). Top: 10× magnification; bottom:60× magnification. (B,C) Quantification of ASC specks over time in ASC–citrine BMDMs adhered to E-selectin (E-sel), IgG2b, or P-selectin (P-sel) following treatment with LPS + VCR (B) or LPS + nigericin (C). E-selectin adhesion significantly increased ASC speck formation compared to controls. This effect was fully blocked by pre-treatment with the NLRP3 inhibitor MCC950 (10 µM) and was mirrored by IL-1β release (D). (E) Adhesion to E-selectin significantly increased the proportion of active p65 NF-κB across all treatment conditions (saline, vincristine, LPS, and LPS + vincristine) compared with IgG2b controls, whereas P-selectin did not induce comparable NF-κB activation. (F) Intraplantar (i.pl.) injection of E-selectin causes IL-1β-mediated mechanical hypersensitivity that is prevented by the IL-1 receptor antagonist anakinra (100 mg/kg; ip.). (G) Intraplantar administration of anakinra or saline prior to PBS or IgG2b i.pl. injection does not cause mechanical hypersensitivity. (H) Intraplantar injection of BMDMs treated with LPS + VCR and adhered to E-selectin induced significantly greater mechanical hypersensitivity compared to IgG2b-adhered controls. Injection of untreated BMDMs adhered to either E-sel or IgG2b had no effect. Statistical significance (#: P < 0.05) was determined using repeated measures two-way ANOVA (n = 6). Statistical significance (n=6) was determined by repeated-measures two-way (C,F,G,H) or one-way ANOVA (D,E) with Dunnett’s multiple comparisons test (#P < 0.05).

Representative immunohistochemical images of F4/80⁺ area in dorsal root ganglia and sciatic nerve 25 h post-vincristine administration.

Mice received two intraperitoneal (i.p.) injections of antibodies targeting E-selectin (E-sel Ab), P-selectin (P-sel Ab), VCAM-1 (VCAM-1 Ab), ICAM-1 (ICAM-1 Ab), and PECAM-1 (PECAM-1 Ab), or corresponding isotype control antibodies (IgG2B, IgG1, IgG2A(TTP)) (10 mg/kg; i.p.) at two time points (t -24h, t -1h). This was followed by a single vincristine injection (0.5 mg/kg, i.p., time point t 0h). Shown are representative pictures of F4/80⁺ area in dorsal root ganglia (A-J) and sciatic nerve (K-T) in a saline control group (A; K) and following pretreatment with isotype control antibodies (B-E; L-O) or adhesion molecule-inhibiting antibodies (F-J;P-T). Scale bar = 20 µm for DRGs or 10 µm for SN. Examples of F4/80⁺ area (brown stain) indicated by arrowheads.

Vincristine induces a stress and injury transcriptional programme in dorsal root ganglia without upregulating E-selectin.

(A–C) Volcano plots of pseudobulk differential gene expression in dorsal root ganglia (DRG). (A) C57BL6/J saline versus vincristine (VCR) reveals induction of injury- and excitability-associated genes (e.g. Map1b, Hcn2) and downregulation of haemoglobin genes (Hba-a1, Hba-a2, Hbb-bs). (B) C57BL6/J (VCR) versus Sele⁻/⁻(VCR) and (C) C57BL6/J (VCR) versus C57BL6/J E-selectin antibody pre-treatment (Esel-ab + VCR) identify stress-related transcripts (e.g. Hspb1, Scand1) associated with E-selectin perturbation. (D–E) Circulating soluble E-selectin plasma protein levels in C57BL6/J mice treated with vincristine (0.5 mg.kg, i.p., once) and Sele transcript expression in HUVEC cells treated with vincristine for 24 hours, show no induction of Sele expression by vincristine, while Tnf serves as a positive control (ns, not significant). (F) UMAP projections of cell types across conditions (saline, VCR, Sele⁻/⁻ VCR, Esel-ab + VCR) demonstrate preservation of major neuronal and non-neuronal cell populations. Abbreviations: VCR: vincristine; Sele⁻/⁻: E-selectin–deficient mice; Esel-ab: E-selectin–blocking antibody; DRG: dorsal root ganglion; TNF: tumour necrosis factor; Schwann_M: myelinating Schwann cells; Calca⁺Bmpr1b and Calca⁺Smr2: peptidergic sensory neuron subtypes; Mrgpra3⁺Trpv1: nociceptive/pruriceptive neurons; Mrgprd: non-peptidergic sensory neurons; Ntrk3^high/+Ntrk2 and Ntrk3^low/+Ntrk2: sensory neuron subsets; Pvalb: parvalbumin-expressing interneurons; Sst: somatostatin-expressing neurons; Th: tyrosine hydroxylase–expressing neurons; Immune: aggregated immune cell populations; sE-sel: soluble E-selectin.

Vincristine and E-selectin perturbation reshape growth factor and neuromodulatory communication networks in dorsal root ganglia.

CellChat analysis was used to infer ligand–receptor–mediated communication networks in dorsal root ganglia (DRG) following vincristine treatment, highlighting pathway-specific sender, receiver, mediator and influencer roles across neuronal and non-neuronal cell populations. (A–E) Growth factor and neuromodulatory signalling pathways in C57BL6/J mice treated with vincristine (VCR). Heatmaps depict the relative contribution of each annotated cell type as senders, receivers, mediators or influencers for KIT (A), NRG (B), NT (C), NTS (D) and FGF (E) pathways, with darker green indicating higher importance scores. (F–J) Corresponding analyses in Sele⁻/⁻ mice treated with vincristine (VCR) show preservation of KIT, NRG, NT, NTS and FGF pathways but redistribution of communication roles across neuronal, glial and stromal populations. (K–S) Analyses in C57BL6/J mice treated with vincristine following E-selectin antibody pre-treatment (Esel-ab + VCR). Heatmaps illustrate altered sender, receiver, mediator and influencer contributions for SPP1 (K), GALECTIN (L), PTN (M), SEMA3 (N), KIT (O), NRG (P), NT (Q), NTS (R) and FGF (S) pathways, highlighting further redistribution of signalling roles relative to genetic E-selectin deletion. Across all panels, rows indicate inferred communication roles (sender, receiver, mediator, influencer) and columns represent annotated DRG cell populations; colour intensity reflects relative importance within each pathway. VCR: vincristine; Sele⁻/⁻: E-selectin–deficient mice; DRG: dorsal root ganglion; Schwann_M: myelinating Schwann cells; Calca⁺Bmpr1b and Calca⁺Smr2: peptidergic sensory neuron subtypes expressing Calca with Bmpr1b or Smr2, respectively; Mrgprd: non-peptidergic sensory neurons; Mrgpra3⁺Trpv1: nociceptive/pruriceptive neurons; Ntrk3^high/+Ntrk2 and Ntrk3^low/+Ntrk2: sensory neuron subsets defined by relative Ntrk3 and Ntrk2 expression; Pvalb: parvalbumin-expressing interneurons; Sst: somatostatin-expressing neurons; Th: tyrosine hydroxylase–expressing neurons; Immune: aggregated immune cell populations; KIT: stem cell factor receptor signalling; NRG: neuregulin signalling; NT: neurotrophin signalling; NTS: neurotensin signalling; FGF: fibroblast growth factor signalling; SPP1: osteopontin signalling; PTN: pleiotrophin signalling; SEMA3: semaphorin 3 signalling.

(A) Control groups treated with liposomes (LIP), clodronate liposomes (CLD), saline or IgG2b did not develop mechanical hypersensitivity. (B) Representative images of F4/80⁺ immunohistochemistry in footpad skin of mice pretreated with CLD or LIP and injected with i.pl. E-selectin. Statistical significance (#: P < 0.05) was determined using repeated measures two-way ANOVA (n = 6-9).

E-selectin enhances ASC speck formation and IL-1β release in vincristine- or nigericin-treated BMDMs.

(A) No significant ASC speck formation was observed in E-selectin–adhered BMDMs treated with LPS, MCC950, nigericin, vincristine (VCR), or saline alone. (B,C) The NLRP3 inhibitor MCC950 (10 µM) prevents ASC speck formation (B) and IL-1β release (C) induced by treatment with lipopolysaccharide (LPS, 10ng/ml) + nigericin (5 µM). (D–E) IL-1β release from BMDMs treated with LPS + doxorubicin (DOX, 50 µM) (D) or LPS + cyclophosphamide (CYC, 100 µM) (E) adhered to E-sel, IgG2b, or P-sel. No significant differences were observed across conditions. Statistical analysis: two-way ANOVA (A-B), one-way (C,D,E) (#P < 0.05); n≥3.